Simple Attenuators - Design And Testing

[Marcomel79]

@Marcomel79
I think you're going to find that all tone is well retained at all attenuation settings with what you currently have. What will come back, when you're coil arrives, is the tactile feel and response, making it all seem even a bit more natural. There is little as pleasing as enjoying your amp at it's "Sweet Spot" without having family and friends angrily screaming "TURN IT DOWN!" It will be interesting to see how you finish up the external cosmetics!
Great Job On Another Succesful Build & Happy New Year!
Gene

Gene, im beside myself! I tried it today with my superlead clone and its amazing! I can now have the high treble channel on 6/7 without the wife filing for divorce!
I was playing for an hour earlier and the box barely got warm. Also, i have not drilled the lid vents yet... i cannot recommend it enough. I might build one each for a couple friends of mine.

Look forward to getting and installing the coil. Also thank you so much for all the help, this was a team effort!

Happy new year to you too!

 

[Gene Ballzz]

Gene, im beside myself! I tried it today with my superlead clone and its amazing! I can now have the high treble channel on 6/7 without the wife filing for divorce!
I was playing for an hour earlier and the box barely got warm. Also, i have not drilled the lid vents yet... i cannot recommend it enough. I might build one each for a couple friends of mine.

Look forward to getting and installing the coil. Also thank you so much for all the help, this was a team effort!

Happy new year to you too!

Yeah, these things are cheap and easy enough to build (especially after the first one) that I'm working towards having one to install in each and every amp I own. The physical shape and size may need a little tweaking for each amp, especially the head models. The most tedious part of the build is ventilation drilling! I'm trying to envision a more convenient solution! Maybe using two panels of perforated aluminum or stainless sheet and cutting the top and bottom to mount that perforated sheet to, from the inside? Something like this:

https://www.amazon.com/Stainless-Pe...ocphy=9030801&hvtargid=pla-669006187633&psc=1

OR THIS:

https://www.metalsdepot.com/aluminu...DCogjQhPTLU_F8QsOPP8D1GqeA8TB2IBoCsooQAvD_BwE

AND A GOGGLE SEARCH:

perforated aluminum sheet

The extruded aluminum boxes I've previously used, as well as the die-cast boxes from Hammond are fairly easily worked with a router, so you could easily cut a rounded corner, rectangular hole, router/round the edges over and mount the panel from the inside, on top and bottom. It would also provide a wide array of mounting holes for the components and have the screw heads recessed in those cutouts! It would still provide the heat sinking of attaching to the aluminum, as well as having massive amounts of flow through, convection ventilation! I'll try a probably crude drawing, when I get the chance!

Just Thinkin'
Gene

 

[diego_cl]

Yeah, these things are cheap and easy enough to build (especially after the first one) that I'm working towards having one to install in each and every amp I own. The physical shape and size may need a little tweaking for each amp, especially the head models. The most tedious part of the build is ventilation drilling! I'm trying to envision a more convenient solution! Maybe using two panels of perforated aluminum or stainless sheet and cutting the top and bottom to mount that perforated sheet to, from the inside? Something like this:

https://www.amazon.com/Stainless-Pe...ocphy=9030801&hvtargid=pla-669006187633&psc=1

OR THIS:

https://www.metalsdepot.com/aluminu...DCogjQhPTLU_F8QsOPP8D1GqeA8TB2IBoCsooQAvD_BwE

AND A GOGGLE SEARCH:

perforated aluminum sheet

The extruded aluminum boxes I've previously used, as well as the die-cast boxes from Hammond are fairly easily worked with a router, so you could easily cut a rounded cornoer, rectangular hole, router/round the edges over and mount the panel from the inside, on top and bottom. It would also provide a wide array of mounting holes for the components and have the screw heads recessed in those cutouts! It would still provide the heat sinking of attaching to the aluminum, as well as having massive amounts of flow through, convection ventilation! I'll try a probably crude drawing, when I get the chance!

Just Thinkin'
Gene

I've been thinking about something similar, because drilling is too much work.

For the hammond cases I think one alternative could be to use a jigsaw and cut out material from the sides of the case. Another option could be using longer bolts with a spacer, so there's a gap between the case and the cover. But I think both options would look terrible.

I want to build a 100W attenuator using a thick sheet of aluminum (6mm) as a heatsink. And I want to place that sheet inside a case made of bent steel and wood. I don't have a hammond case big enough for that job.

 

[Amped eMind]

This thread is pure gold, as is the heart of John, Gene and all the numerous and generous posters who contribute their knowledge and ideas to the benefit of the community!!!

I read the whole thread a while back before subscribing and probably will skim through again in the near future. I got the parts for an M3 build (go big or go home?) but have to get through a number of unrelated projects first so it is still a few weeks out.

Here's wishing you all a wonderful New Year!

 

[XTRXTR]

@Gene Ballzz I have not built one of these, I have read much of this thread over time and I know the design is proved out. Its the final things like layout and materials that are still being tweaked and developed for a variety of specific use types. What about an actual heat sink design. I am not a metallurgist but I think the volume and density of the heat sink make a difference. Plainly I mean the more metal used the less air flow is required. I would have to double check some thermodynamics to verify that absolutely. If its true a design where the heat sink has enough volume that you could rout out the inside enough for the RL layout and then use a simple box as a cover that only houses the jacks and switches on the walls. Or just place the layout directly on the flat side of the heat sink and cover with the box. Hmm you may have to use blade connectors to remove or replace the box. Well I'm just throwing out ideas.

@JohnH This is a really cool device and I like that everyone can do their own layout to fit the need.

 

[Gene Ballzz]

@XTRXTR
I'd love to see some layman's results of your thermodynaic research and I'll bet others would too! I'm gonna make a guess that it will actually come down to a balancing act. Meaning that as the heat sinking increases the need for ventilation decreases and vice/versa, but only to a certain point. The other part of that guess is that just one or the other will be lacking, to some degree? I think one important factor is to keep the unit as compact as possible, while avoiding the inconvenience of providing power for a fan. Keeping the price down is also a big consideration!

And yes indeed, the drilling, while fairly simple and easy is a time consuming, tedious, ROYAL P.I.T.A.

Keep On Thinkin'
Gene

 

[XTRXTR]

There are people that have made Play stations without a fan because they designed the chassis itself to be a heat sink.

The resistor sizes are determined by the dissipation needed, so the increased size of the heat sink could offset the size of the resistor power rating a little. Anyway, just trying to throw in some ideas.

 

[JohnH]

I ran some tests on my one, soon after I built it. Mine is in a Hammond=style 1590 case (see post 1). I'm into the thermodynamics based on my day job, so it was another interesting aspect for me to over-analyse!

My test was to feed in dc , since that way I knew from voltage and resistance exactly what the power was, which came out at 7.2W continuous. It started to heat up and I made measurements over time with an infra-red thermometer. As heavy as it was, it took a long time to heat up. It went from 21C to 28C in 20 minutes. It was at 31C in about 40 minutes, and after 75 minutes got to about 33C. Id expect that with more power, it would take a similarly long tome o reach max temperature, (which would be higher of course)

So nothing happens fast with a thick Al case, and the power producing heat can be considered as an average over maybe 20 to 30 minutes.

Next, to understand the thermodynamics: There are 4 basic mechanisms in play:

Conductivity: This is the parameter defining heat flow through the solid case. Al is many times better than any steel and the thicker and more continuous the metal is, the easier the heat flows away from the heat source and spreads out through the case. The thermal paste bridges a potential bottle neck in the flow. (btw, id not be too keen on bolting the stage 1 resistors to a perforated sheet, they are not very thick and the flow of heat through the material is constrained by all the holes)

Thermal capacity: Determines how fast it heats up with a given heat input. The slow temperature response is due to the high thermal capacity of the relatively heavy case and resistors.

That leaves 2 other parameters which set the cooling, and they are radiation and convection. Both are dependent on the heated surface area. Convection is cooling due to the flow of fluid (air in this case) and the temperature difference between object and air. Radiation is electromagnetic radiation through space. It depends (very interestingly!) on the 4th power of absolute temperature, and radiative cooling is the balance between heat radiated out and that received in from the surroundings.

So, all of that has known formulae applicable, and i had a go at assessing it. For my test, based on what I could work out I figured that about 60% of the cooling was by radiation and 40% by convection.

We can use such a test to calibrate the maths. Ive never been able to get my attenuator really hot, running probably at around 20-30W on my 50W VM, but semi clean and not a continuous thrash. Lets say over a decent session, my average power is 25W, then Id predict about 60C, which is fine. But if you drive to 50W on average, Id predict 90C, which is too hot to touch. Hence possible use of fans for use with larger amps, which can increase convective cooling by a factor of 2.5 to 3.

 

[Gene Ballzz]

@JohnH
Thanks for sharing that! So my take away is that about 75°C is the absolute maximum we could tolerate seeing, over extended periods of use? And it would likely take close to an hour or so to get there, pounding steadily at 45 to 50 watts, with a well ventilated 3mm thick case? And if it levels off at that 70-75 degrees, we're probably OK? the wire I use is rated for 105°C and the switches are rated for 85°C. I'd guess that is pushing the limits a bit!
Just Spitballin'?
Gene

 

[JohnH]

I reckon that's about right. I don't think we are close to electrical problems there, but you don't want to get burnt! As a reference, a metal pot of boiling water at 100C is too hot to touch even momentarily. I reckon if you can touch the hottest parts for a couple of seconds, then its ok.

 

[JohnH]

i looked back again at my heat tests and worked a bit more excel....

I think for my tests, its more like 60% convective 40% radiative

Anyway, here are numbers for anyone who cares to pick through it:



At left is my test. The heat is not the same all over the case, so I found the most reasonable set of numbers as an average, noting that all the hot bits are on the case base, if I assumed the base (=top in my case) is at a constant temp and I ignore the lid area.

I also found online, that a standard for the max temp of hot things hat might get touched is 60C. So feeding that in, I can estimate the power that can be dissipated steady-state at 60C. Assuming a dark case, and 20C ambient temp, its about 600W per sqm of the case effective area. Add a fan to improve convection, then its then about 1100W per sqm.

Then I can estimate how much power from an amp will create that. And then what amp will provide that power? The lowest rows above take the average heat power and estimate the amp spec by dividing by a factor to do with how much of the time the amp is driving over say 30 minutes (I put 80%). Then, how hard is the amp driving during that time? from say 80% of max for cleanish/edge of breakup, to 120% for full powerstage overdrive.

So, with no fans, I get to about a 50W amp for a 1590 case if you play cleanish, or 30W if driven. About 50% more for a larger 1550 case, and add a fan and it suggests around a 100W amp driven, using a 1550 case.

All of the above is speculative and guesswork and the parameters are just assumptions based on minimal data, though using some science. Really its just to explore the parameters.

Does anyone have any anecdotal descriptions of attenuator hottness vs amp power, case size and playing style?

 

[donwagar]

on the low end of the scale, mine is the 50W M2 16 ohm, can't remember the case size (it's at our rehearsal space so I can't measure it) but it's a tight fit, and I didn't have room for a lot of vent holes. Running my SV20 on 20W with almost everything dimed (normal channel on about 5 and high treble on about 8, tone stack all in the 8 range. At gigs, I do 3 one hour sets. It barely gets warm.

And it sounds fantastic.

 

[diego_cl]

M2 50W with chinese 1590D case (2.6mm thick) and resistors (1.9mm thick at its base), both from Aliexpress. Amp is a Marshall MA50C (50W).

It doesn't care if volume is at 6 or 12, but after 20 minutes it gets at least over 65°C if it's distorting (with or without booster pedals). Edge of break up makes the case hot, but not over 65°C.

I'm worried about the inductor enamel. I have to improve the vent holes and I want to put a dedicated heat sink over R1.

I don't have an IR thermometer. I'm just estimating the temperature by using my experience with measured hot water.

 

[JohnH]

Thanks @donwagar and @diego_cl both those descriptions fit in pretty well with my theoretical hack described above.

 

[Gene Ballzz]

Hey Folks,
I just thought of a stupidly simple upgrade for the cosmetic aestheics of our builds. Given the thickness of the enclosures we use, simply utilize flat head machine screws/bolts for mounting the resistors and counter sink them into the case for a much cleaner look!

Simply Screwin'
Gene

 

[Marcomel79]

Hey Folks,
I just thought of a stupidly simple upgrade for the cosmetic aestheics of our builds. Given the thickness of the enclosures we use, simply utilize flat head machine screws/bolts for mounting the resistors and counter sink them into the case for a much cleaner look!

Simply Screwin'
Gene
View attachment 100126

Thats a very good idea Gene!! I purposely left my screws at the bottom so i wouldnt see them. But if it cant be helped and they are visible, thats a very neat idea.

 

[Gene Ballzz]

I have been drilling and tapping the holes in the boxes and then cutting the screws to length, but this is a quite tedious and often "iffy" process! Sometimes fitting a nut on the inside may be a little tough on some resistor brands, but it seems that the holes in the resistors could get redrilled to a slightly larger size, for wiggle room. Alternatively, if the holes are a consistent size, simply tap threads into the resistors, avoiding nuts altogether? I don't know why I didn't think of this sooner!

Of course, its kinda hard to avoid "nuts" like me, in this thread!

Still Screwin'
Gene

 

[Amped eMind]

Speaking of methods of attachment of the resistors, do you guys see many cons on using rivets instead of screws and nuts? Rivets are easy and fast to install and provide a rather neat look (though maybe not as neat as flush screw heads as Gene proposes).

The resistors are not likely to need to be removed and the rivets could always be drilled out if the need really arose.

 

[Gene Ballzz]

@Amped eMind
I see no issue with pop rivets, as long as they actually clamp the resistor firmly to the case, instead of leaving it "standing off" a little bit, as I've sometimes seen rivets do! I would think it most appropriate to firmly clamp each resistor in place before applying the rivets and then make sure the rivet head falls out or gets knoked out, so that it doesn't end up inadvertently falling out to short out the circuitry?
Just Clampin'
Gene

 

[Amped eMind]

Yes, I'll make sure the resistors are held firmly in place and that the pop rivets are tight after they snap. Thanks!